Digital images which can be captured by scanners or digital cameras, or produced by a computing process, have to be rendered so that they can be properly viewed on a display. The resulting digital images are often referred to as being in a “rendered” image space, such as sRGB, where the relationship between the image code values and the scene luminance values is very non-linear, as depicted in FIG. 8. Digital images, whether obtained from scanned film or from digital cameras, often contain exposure errors, where the key subjects of the picture are lighter or darker than desired by the user, due to imperfect exposure determination algorithms in the digital camera or film scanner which created the digital image. Many imaging applications, such as Adobe Photoshop or Kodak PictureEasy, permit the user to adjust the “brightness”, “contrast”, and/or “gamma” of the image by sliding using one or more “sliders” controlled by a mouse. Each control adjusts one of the slope, x-intercept, or exponential function applied to the image, typically using a look-up table, in order to modify the tone reproduction of the image when it is displayed or printed. However, because they do not directly adjust the scene exposure, they do not properly compensate for camera exposure errors. While it is possible to somewhat improve the image using such controls, it is extremely difficult for unskilled users to determine how to best set these multiple controls.
Some applications, such as Adobe Photoshop, also permit the user to modify a “lookup table” by inputting a curve of any arbitrary shape. However, only a skilled user is able to determine what curve shape provides the best image, and only then using a tedious trial-and-error process.